Density Functional Theory Investigations on the Structure and

Dec 15, 2011 - INTRODUCTION. Shrinking fossil fuel resources and environmental crises are major problems in the present world, which necessitate us to...
0 downloads 0 Views 4MB Size
ARTICLE pubs.acs.org/JPCB

Density Functional Theory Investigations on the Structure and Dissolution Mechanisms for Cellobiose and Xylan in an Ionic Liquid: Gas Phase and Cluster Calculations Rajdeep Singh Payal, R. Bharath, Ganga Periyasamy, and S. Balasubramanian* Chemistry and Physics of Materials Unit, Jawaharlal Nehru Centre for Advance Scientific Research, Jakkur, Bangalore 560064, India

bS Supporting Information ABSTRACT: Density functional theory (DFT) calculations have been carried out for cellobiose and xylan chosen as models for cellulose and hemicellulose, respectively, in gas phase, implicit and explicit solvent (water, methanol, and the ionic liquid, 1,3-dimethylimidazolium acetate) media using plane wave and atom centered basis set approaches in order to find out lowest energy conformers and configurations. Geometry, vibrational properties, and 1H and 13C NMR chemical shift values have been discussed under all three conditions. Calculations predict that inter- and intramolecular hydrogen bonding play an important role in the dissolution processes. In the gas phase and in implicit solvent, the antianti conformer of cellobiose and the antisyn conformer of xylan are the most stable due to the formation of a large number of intramolecular hydrogen bonds. However, in the cluster calculations containing ion pairs of the ionic liquid (IL) surrounding the cellulosic units, the antisyn conformer of cellobiose is more stable as intramolecular hydrogen bonds are substituted by intermolecular ones formed with the ions of the IL. The complexes of cellobiose (or of xylan) with the ions of the ionic liquid are stable with large negative binding energies ranging between 21 and 55 kcal mol1. The predicted 1H NMR values of the lowest energy cellobiose conformers are in good agreement with the experimental value. Xylan binds stronger with the IL than cellobiose does by 20 kcal mol1. Furthermore, the two pentose rings in xylan are rotated by 60° to each other in contrast to their coplanarity in cellobiose, which can explain the higher solubility and the amorphous nature of hemicellulose in ionic liquids. The fewer number of hydroxyl groups in xylan (relative to cellobiose) does not affect the number of cations present in its first solvation shell while the number of anions is reduced.

1. INTRODUCTION Shrinking fossil fuel resources and environmental crises are major problems in the present world, which necessitate us to reevaluate the efficient utilization of energy sources and find ecofriendly alternatives, such as, biorenewables.13 Lignocellulosic material which is most abundant in wood/plant biomass is an example of such a resource. It chiefly contains lignin (2030%), cellulose (4050%), and hemicelluloses (2030%).46 Dissolving these organic polymers and breaking them into intermediate compounds that can be converted to biofuel (ethanol)/ artificial silk remains a difficult task. Till now, only 40% of the ethanol content available in cellulosic feedstock is extracted effectively.3 Cellulose and hemicellulose are polysaccharides, present in plant cell wall with β-(1f4) linked backbones in an equatorial configuration.79 Although both are polysaccharides, cellulose is a crystalline polymer with a highly ordered structure whereas hemicellulose is amorphous with a random structure, and many of them contains xylan backbone. In cellulose, cellobiose units are connected to one another via strong inter- and intramolecular hydrogen bonding networks (see Figure 1a) thus providing it and the cell wall with high chemical and mechanical stability. However, amorphous hemicellulose stabilizes the cell wall by r 2011 American Chemical Society

interacting with neighboring hemicellulose, cellulose, or lignin units (Figure 1b). Many research groups have tried to dissolve lignocellulosic material in common organic solvents at various physical conditions.916 Some of the solvents such as N-methylmorpholine-Noxide monohydrate, N,N-dimethylacetamide/LiCl, and LiClO4 3 3H2O have been shown to dissolve cellulose up to 1520% of weight of lignocellulosic biomass. However, these solvents have disadvantages such as toxicity, instability, or high cost.917 To overcome these issues, Rogers et al.18 used room temperature ionic liquids (RTIL) as a solvent to dissolve cellulose. These ionic liquids are organic salts that have low melting temperatures ( O2H2 ≈ O20 H20 ≈ O3H3 ≈ O10 H10 > O6H6 ≈ O60 H60 ≈ O4H4 and also reflected in the bond distances as shown in Figure 4. Similarly, two types of OH stretching modes were observed for the antisyn conformer with respect to their H-bond formation ability. Room temperature 1H and 13C NMR data of synthetic cellobiose analogues and of [C2mim][OAc] have been reported in DMSO-d6 solvent at different mole fractions.49,50 However, chemical shift values were discussed only for specific atoms due to experimental difficulties. It would thus be useful for us to validate our computational work against these experimental data. The calculated 1H NMR chemical shift values of 25 ppm for the H-bonded OH and 13C NMR values 5595 ppm for the anti anti conformer are comparable to the experimental 1H (35 ppm) and 13C NMR values (4991 ppm) of cellobiose synthetic analogues. In contrast, the corresponding values for the antisyn conformer are 1H, 1.52.5 ppm and 13C, 52 90 ppm (see Table S1 of the Supporting Information).47,73 Minor differences in the chemical shift value are likely due to differences in the functional groups between the synthetic analogues and cellobiose and also due to the absence of a solvent environment in these calculations. The presence of one extra H-bond in antianti conformer (O30 H30 3 3 3 O6) shifted the H30 1H NMR value to the downfield (4.65 ppm) than in the antisyn conformer (2.07 ppm). Furthermore, the computed energy difference of 4.53 kcal mol1 between the antianti and antisyn conformers at the B3LYP/6-31++G(d, p) level is in good agreement with the value of 5 kcal mol1 as reported through MP2/6-31G(d) calculations in the gas phase.43 These data further validate the methods adopted here. 3.2. Gas Phase Structure of Xylan. The lowest energy conformers for xylan are analyzed based on the Ω0 and Φ0 dihedral angles. In contrast to cellobiose, the two pentose rings in xylan are oriented at 70° to each other. This prevents the formation of a large number of intramolecular H bonds as in cellobiose. The antisyn conformers 1 and 2 (shown in Figure 5) are lower in energy than

the others (antianti and syn-syn). The antisyn conformers differ from each other in the orientation of the OH groups, which leads to different number of intramolecular H bonds. The larger number of intramolecular H bonds in the antisyn conformer C2 (panel b of Figure 5) stabilizes it by 3.22 kcal mol1 (see Table 2) than conformer C1 (panel a of Figure 5). Similar to the case of cellobiose, the presence of H-bonding interactions is reflected in the OH stretching frequencies and 1H NMR values. The presence of both free (i.e., OH group without any H bonding interaction) as well as H-bonded OH groups in the conformers leads to a large spread of OH stretching frequency ranging from 3770 cm1 to 3820 cm1 (see Table 3). This is also observed in the values of 1H NMR chemical shift which is in the range of 1.87 2.28 ppm for the H-bonded OH group. Extra H-bond in antisyn conformer 2 shifts the 1H NMR value of the H20 (1.27) atom to downfield more than the antisyn conformer 1 H20 (0.79). Compared to the values for free OH, these are shifted downfield by 1.01.6 ppm. In addition, the structural differences between cellobiose and xylan are reflected in the 13C NMR chemical shift values as shown in Table S1.

4. IMPLICIT SOLVATION While gas phase calculations provide details of the structure and hydrogen bonding in the cellulosic units, it will be more appropriate to compare experimental data to a system in which such units are solvated at the same thermodynamic conditions as in experiments. With advancements in simulation methods and computing environments, this is an achievable aim to a large extent through the method of ab initio molecular dynamics simulations. Pending such simulations which we plan to carry out in future, in the current work we have studied the effect of solvation of cellulosic units through two “gas phase” quantum techniques: (i) using a dielectric continuum model and (ii) cluster calculations of cellulosic units surrounded by layer(s) of solvent molecules. In this section, we discuss results of the former. 4.1. Implicit Solvation Effects on Cellobiose. Four lowest energy conformers of cellobiose obtained from plane wave calculations in the gas phase were further optimized within dielectric media using the atom centered basis set approach (Gaussian 03).57 The inclusion of the dielectric medium did not affect the order of the stability of the conformers. However, it reduced the energy difference between the stable conformers (see Figures 4 and 5 for the gas phase structures) by 2.56 kcal mol1 (IL), 2.93 kcal mol1 836

dx.doi.org/10.1021/jp207989w |J. Phys. Chem. B 2012, 116, 833–840

The Journal of Physical Chemistry B

ARTICLE

Figure 6. Mapping of the molecular electrostatic potential (MEP) on an isocontour of the electron density taken to be 0.04 e bohr3 for (a) water, (b) methanol, and (c) an ion pair, [C1mim][OAc].

Figure 7. Computed lowest energy configurations of cellobiose in (a) water and (b) methanol clusters. Nine inter- and four intramolecular hydrogen bonds are seen. Various structural parameters are also shown.

Figure 8. Computed lowest energy configurations of cellobiose surrounded by ion pairs of [C1mim][OAc] ionic liquid. Intermolecular H bonds between cellobiose and the ions as well as relevant structural parameters are shown.

(water), and 3.02 kcal mol1 (methanol) (see Table 2). Increasing the dielectric constant decreases the energy difference between the antianti and antisyn conformer by the following amounts: for IL (1.97 kcal mol1), methanol (1.60 kcal mol1), and for water (1.27 kcal mol1). The hydrogen bond length is seen to correspondingly increase by 0.010.03 Å. The subtle change in the interaction strength is also reflected in the OH stretching mode and in the 1H NMR chemical shielding values (see Tables 3 and S1). The OH stretching mode of antianti cellobiose in IL is redshifted by about 200 cm1 relative to the gas phase value. Implicit solvation also changes the 1H NMR chemical shift values of the hydroxyl group upfield by 0.20.4 ppm than in the gas phase due to the weakening of H bonds. The computed 1H and 13C NMR chemical shift values for the antianti conformer in the implicit solvation media are in good agreement with experimental values. This provides further evidence for the stability of the antianti conformer of cellobiose molecule. 4.2. Implicit Solvation Effect on Xylan. Similar to cellobiose, implicit solvation did not affect the conformer stability order of xylan. However, the energy difference between the two conformers was reduced by 0.69 kcal mol1 in the implicit presence of the ionic liquid. For both conformers of xylan, the H-bonded OH stretching frequencies were red-shifted (37803830 cm1 for conformer 1 and 37603800 cm1 for conformer 2) than for the free OH group (38103840 cm1 for conformer 1 and 38103860 cm1 for conformer 2) (see Table 3). The polarization effects due to intramolecular H bonds are also clearly reflected in the 1H NMR (2.913.74 ppm) and 13C NMR (41.93 105.04 ppm) chemical shielding values, which is shifted upfield by 0.82.9 ppm compared with their gas phase values (see Table S1). The implicit solvations studies suggest that a solvent with high polarization could dilute intermolecular H bonds of cellobiose and xylan. To verify this, calculations with explicit solvent molecules need to be carried out. This is the object of the next section.

5. EXPLICIT SOLVATION EFFECTS An understanding of explicit solvent effects on the energetics and structure of cellobiose and xylan units requires an appreciation of the electrostatics of the molecules involved. Compared with water (oxygen: 0.720 e; hydrogen: 0.36 e) and methanol (oxygen: 0.541 e, hydrogen: 0.340 e), the acetate oxygen (0.751 e) and the acidic hydrogen of the imidazolium cation (0.527 e) have high negative and positive Mulliken atomic charges respectively. This suggests the higher ability of IL to donate or accept H bonds relative to water or methanol. The difference in charge density between the three solvents is clearly seen in the molecular electrostatic potential maps too (Figure 6). 5.1. Explicit Solvation Effects: Cellobiose. The computed lowest energy conformers in the gas phase for cellobiose were also optimized in the presence of explicit solvent environment. A layer of solvent molecules were added at arbitrary locations around the cellobiose molecule, as an initial configuration for geometry optimization runs. In the optimized configuration, water and methanol molecules form intermolecular H bonds with cellobiose as shown in Figure 7. However, this solvation does not completely break the intramolecular H-bonding between cellobiose moieties. The calculated negative BE values ranging between 10 and 20 kcal mol1 reflect the stability of water/methanol solvated cellobiose cluster. The antianti conformer is highly stable in both water and methanol media than the antisyn conformer, similar to its stability in the gas and implicit solvent phases. We notice that the small size of the water/methanol molecules allows for all the nineteen solvent molecules to be present in the first coordination shell of cellobiose. Further, intermolecular H-bonding is seen to weaken the intramolecular H bonds, which can decrease the energy 837

dx.doi.org/10.1021/jp207989w |J. Phys. Chem. B 2012, 116, 833–840

The Journal of Physical Chemistry B

ARTICLE

Table 4. Computed Energy Differences (ΔE) between the Most Stable Configurations in Explicit IL Solvent Mediuma molecules cellobiose

xylan

ΔE (kcal mol1)

IL clusters

binding energy (kcal mol1)

configuration 1 (antisyn conformer)

0.0

21.10

configuration 2 (antianti conformer)

6.02

11.05

configuration 3 (antisyn conformer)

11.22

9.98

configuration 4 (antisyn conformer)

13.13

8.07

configuration 1 (antisyn conformer 1)

0.0

55.39

configuration 2 (antisyn conformer 2)

16.52

35.26

Binding energy (BE) = EIL‑cellobiose/xylan cluster  (Ecellobiose/xylan + NEIL pairs), where N denotes the number of solvent ion pairs (N = 7). Note that the ZPE corrections are not included in the ΔE, BE values. a

difference between the antianti and antisyn conformers by 0.20.3 kcal mol1 (see Table S2). However, the effect on cellobiose due to the explicit solvation by ion pairs of an IL is vastly different. In particular, the larger size of the cation and the anion prevents the direct, proximal interaction between every ion and the OH group of cellobiose. This is clearly reflected in the computed minimum energy geometries of cellobiose which are shown in Figure 8. Optimization of cellobiose solvated by ion pairs of [C1mim] [OAc] leads to four low energy configurations, where the cellobiose unit is present in antisyn configuration 1 (Figure 8a), antianti configuration 2 (Figure 8b), antisyn configuration 3 (Figure 8c), and antisyn configuration 4 (Figure 8d). All four configurations have negative binding energies ranging from 8.07 to 21 kcal mol1 (see Table 4), which indicates the strong interaction between the ions and cellobiose. The computed BE values in “IL medium” is larger than the corresponding values for water and methanol media. The high H-donating and accepting nature of ion pairs results in the formation of strong intermolecular H bonds and also breaks all the intramolecular H bonds of cellobiose. Dramatically, these effects swap the order of conformer stability. In the presence of [C1mim][OAc], the antisyn configuration becomes more stable than the antianti conformer by 6 kcal mol1 (see Tables 2 and 4). This is in contrast to the observations on conformer stability found in the gas, implicit solvent, explicit water, and methanol media. Furthermore, in agreement with NMR studies,49,50 both the cation and the anion of the IL are seen to interact with the cellobiose unit. All the seven anions form strong H bonds with the cellobiose solute. However among the seven cations, only four form H bonds with cellobiose. The remaining cations form H bonds with the anions alone. Values of dihedral angles for the four lowest energy conformers studied here are in good agreement with those seen in molecular dynamics simulations. The values seen in MD simulations were (j = 125°, ψ = 50°), (j = 50°, ψ = 100°), and (j = 120°, ψ = 50°).46 The changes from intra- to intermolecular H-bonding network are also clearly reflected in the 1H NMR values of the hydroxyl group as shown in Table S1 of the Supporting Information. Strong intermolecular H-bonding interactions are the reason for larger downfield movement of 1H and 13C chemical shifts by 3.54.6 ppm and 66.580.1 ppm compared to the values observed in gas and implicit solvent media calculations. The computed chemical shift values are comparable with experimental 1 H (3.06.0 ppm) and 13C (68.0103.0 ppm) NMR chemical shift values.49,50 Minor differences could be due to the lack of more solvent layers in the calculations. In the next section, we examine the stability of conformers of xylan in the explicit IL medium.

Figure 9. Computed lowest energy structures of xylan in the IL solvent medium. Intermolecular H bonds between xylan and ion pairs and relevant structural parameters are shown.

5.2. Explicit Solvation Effects on Xylan. Similar to cellobiose, xylan forms intermolecular H-bonding with ions which leads to negative binding energy values of 33 to 51 kcal mol1. The computed binding energy for xylan is higher by 29 kcal mol1 than that for the cellobiose molecule, which is a likely reason for the high solubility of hemicellulose in IL media over that of cellulose.17 Figure 9 shows two of the low energy configurations of xylan “solvated” by ions of the ionic liquid. Again, conformer 2 (Figure 9b) differs from conformer 1 (Figure 9a) in the orientation of the OH group indicated by an arrow in Figure 5. Similar to cellobiose, the xylan molecule also interacts with four cations. This shows that the fewer number of hydroxyl groups on xylan does not affect its interactions with the cation. However the number of anions forming hydrogen bonds with xylan is reduced to a value of six from the value of seven that was observed for cellobiose. This difference could be a minor one and its importance (or otherwise) can be checked only by doing simulations of the solute in explicit bulk solvent. Further, six out of seven anions of the IL form H bonds with xylan. The intermolecular H bonds between the ions and xylan are shorter by 0.10.3 Å than the corresponding distances between the ions and cellobiose., There are no intramolecular hydrogen bonds in both the configurations of solvated xylan. Configuration 1 contains ten intermolecular hydrogen bonds (i.e., xylan forming H bonds either with the cation or the anion) stabilizing it over configuration 2 in which only seven intermolecular H bonds are observed. The reduction in the number of H bonds in conformer 2 is due to the orientation of the OH group indicated by an arrow in Figure 5. These results suggest that the cation in the IL could also be involved in the solvation by the formation of intermolecular H bonds with the cellulosic units, similar to those formed by the anion. In addition, the xylan molecules 838

dx.doi.org/10.1021/jp207989w |J. Phys. Chem. B 2012, 116, 833–840

The Journal of Physical Chemistry B

ARTICLE

maintain the 60° orientation between the pental rings as in gas phase, which might be the reason for its amorphous nature.

6. CONCLUSIONS We have carried out quantum chemical calculations of cellobiose and xylan in order to obtain a microscopic understanding of their solubility in ionic liquids. The calculations, using density functional theory, have been performed in (a) gas phase, (b) under implicit solvent conditions, and (c) with a layer of explicit solvent molecules. The solvent molecules considered were water, methanol, and 1,3-dimethylimidazolium acetate, [C1mim][OAc]. Our study shows that the conformational stability of cellobiose and xylan are highly dependent upon the number of inter- and intramolecular H bonds. In the gas phase, the moiety with large number of intramolecular H bonds (the antianti conformer of cellobiose and the antisyn conformer 2 of xylan) is most stable. The structural features were reflected in the calculated NMR chemical shifts and vibrational frequencies. The energy differences between the conformers are within 56 kcal mol1 which is within the capability of the solvent to induce a conformational change. The polarization effect due to the dielectric constant of the solvent, captured through continuum calculations, is seen to further reduce the energy difference. The primary cause appears to be the reduction of the H-bond strength; however the conformers follow the same trend in energy as in the gas phase. In addition, our quantum calculations on a cluster of solvent molecules solvating the cellulosic units yield rich information on solutesolvent hydrogen bonding.74,75 While the explicit presence of water and methanol around cellobiose (or xylan) does not change the order of conformational stability, it is seen to partially remove the intramolecular H bonds. Strikingly, all intramolecular H bonds are removed in the explicit IL medium, due to strong solutesolvent H-bonding interaction. As a result, the antisyn conformer of cellobiose and the antisyn conformer 1 of xylan (wherein the glucose units are rotated by about 60°) become more stable than other conformers. Nearly all the anions coat (forming hydrogen bonds) cellobiose exclusively, while only a fraction of the cations do so. This result is in good agreement with experimental NMR studies.49,50 The remaining cations form H bonds with their corresponding anions. The same trend was observed in xylan, where out of the seven ion pairs present in the calculation, four cations and six anions form strong solutesolvent hydrogen bonds. However, the two rings in xylan are oriented at 60° to each other, while in cellobiose they are coplanar. The computed structural parameters of cellobioseIL cluster, especially dihedral values are in good agreement with those seen in earlier molecular dynamics simulations.46 Our studies suggest that both the cation and the anion are responsible for the dissolution of cellobiose and xylan in ionic liquids, which is in good agreement with previous experimental studies.49,50 In addition, our results confirm that the H-bonding interactions between anion and cation with the cellobiose or xylan units contribute significantly in their dissolution. The work also points out the necessity of considering explicit solvent molecules to capture these subtle effects that are so crucial to cellulose solubility in room temperature ionic liquids.

in various environments, and the energetics of the cellobiose methanol and cellobiosewater clusters. Also provided are coordinates of optimized configurations. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT We thank the Department of Science and Technology for support. G.P. thanks DST-Women Scientists Fellowship for funding. ’ REFERENCES (1) Simmons, B. A.; Loque, D.; Blanch, H. W. Genome Biol. 2008, 9, 242. (2) Wynman, C. E. Trends Biotechnol. 2007, 25, 153–157. (3) Richard, T. L. Science 2010, 329, 793–796. (4) Saha, B. C. J. Ind. Microbiol. Biotechnol. 2003, 30, 279–291. (5) Ohno, H.; Fukaya, Y. Chem. Lett. 2009, 38, 2–7. (6) Kennedy, J. F.; Phillips, G. O.; Williams, P. A. Cellulose sources and exploitation: industrial utilization, biotechnology, and physico-chemical properties; Ellis Horwood Ltd.: New York, 1990. (7) Scheller, H. V.; Ulvskov, P. Annu. Rev. Plant. Biol. 2010, 61, 263–289. (8) Pinkert, A.; Marsh, K. N.; Pang, S.; Staiger, M. P. Chem. Rev. 2009, 109, 6712–6728. (9) Nishino, T.; Matsuda, I.; Hirao, K. Macromolecules 2004, 37, 7683–7687. (10) Terbojevich, M.; Cosani, A.; Conio, G.; Ciferri, A.; Bianchi, E. Macromolecules 1985, 18, 640–646. (11) Nishio, Y.; Manley, R. S. J. Macromolecules 1988, 21, 1270–1277. (12) McCormick, C. L.; Dawsey, T. R. Macromolecules 1990, 23, 3606–3610. (13) Williamson, S. L.; Armentrout, R. S.; Porter, R. S.; McCormick, C. L. Macromolecules 1998, 31, 8134–8141. (14) Edgar, K. J.; Buchanan, C. M.; Debenham, J. S.; Rundquist, P. A.; Seiler, B. D.; Shelton, M. C.; Tindall, D. Prog. Polym. Sci. 2001, 26, 1605–1688. (15) Kuang, Q. L.; Zhao, J. C.; Niu, Y. H.; Zhang, J.; Wang, Z. G. J. Phys. Chem. B 2008, 112, 10234–10240. (16) Kubisa, P. Prog. Polym. Sci. 2009, 34, 1333–1347. (17) Zhao, H.; Baker, G. A.; Cowins, J. W. Biotechnol. Prog. 2010, 26. (18) Swatloski, R. P.; Spear, S. K.; Holbrey, J. D.; Rogers, R. D. J. Am. Chem. Soc. 2002, 124, 4974–4975. (19) Welton, T. Chem. Rev. 1999, 99, 2071–2084. (20) Anderson, J. L.; Ding, J.; Welton, T.; Armstrong, D. W. J. Am. Chem. Soc. 2002, 124, 14247–14254. (21) Ludwig, R. ChemPhysChem 2006, 7, 1415–1416. (22) Bhargava, B. L.; Balasubramanian, S.; Klein, M. L. Chem. Commun. 2008, 44, 3339–3351. (23) Maginn, E. J. J. Phys., Condensed Matter 2009, 21, 373101. (24) Fort, D. A.; Swatloski, R. P.; Moyna, P.; Rogers, R. D.; Moyna, G. Chem. Commun. 2006, 714–716. (25) Zavrel, M.; Bross, D.; Funke, M.; B€uchs, J.; Spiess, A. C. Bioresour. Technol. 2009, 100, 2580–2587. (26) Singh, S.; Simmons, B. A.; Vogel, K. P. Biotechnol. Bioeng. 2009, 104, 68–75. (27) Fukaya, Y.; Sugimoto, A.; Ohno, H. Biomacromolecules 2006, 7, 3295–3297. (28) Airong, X.; Wang, J.; Wang, H. Green Chem. 2010, 12, 268–275. (29) Fukaya, Y.; Hayashi, K.; Wadeb, M.; Ohno, H. Green Chem. 2008, 10, 44–46. (30) Sun, N.; Rahman, M.; Qin, Y.; Maxim, M. L.; Rodríguez, H.; Rogers, R. D. Green Chem. 2009, 11, 646–655.

’ ASSOCIATED CONTENT

bS

Supporting Information. Computational details, computed 1H and 13C NMR values of cellobiose and xylan molecules 839

dx.doi.org/10.1021/jp207989w |J. Phys. Chem. B 2012, 116, 833–840

The Journal of Physical Chemistry B

ARTICLE

(66) Hunt, P. A.; Gould, I. R. J. Phys. Chem. A 2006, 110, 2269–2282. (67) Thar, J.; Zahn, S.; Kirchner, B. J. Phys. Chem. B 2008, 112, 1456–1464. (68) London, F. J. Phys. Radium 1937, 8, 397–409. (69) McWeeny, R. Phys. Rev. 1962, 126. (70) Ditchfield, R. Mol. Phys. 1974, 27, 789–807. (71) Wolinski, K.; Hilton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990, 112, 8251–8260. (72) Cheeseman, J. R.; Trucks, G. W.; Keith, T. A.; Frisch, M. J. J. Chem. Phys. 1996, 104, 5497–5509. (73) Pepin, M.; Hubert-Roux, M.; Martin, C.; Guillen, F.; Lange, C.; Gouhier, G. Eur. J. Org. Chem. 2010, 2010, 6366–6371. (74) Shen, T.; Langan, P.; French, A. D.; Johnson, G. P.; Gnanakaran, S. J. Am. Chem. Soc. 2009, 131, 14786–14794. (75) Chundawat, S. P. S; Bellesia, G.; Uppugundla, N.; Sousa, L. C.; Gao, D.; Cheh, A. M.; Agarwal, U. P.; Bianchetti, C. M.; Phillips, G. N.; et al. J. Am. Chem. Soc. 2011, 133, 11163–11174.

(31) Samayam, I. P.; Hanson, B. L.; Langan, P.; Schall, C. A. Biomacromolecules 2011, 12, 3091–3098. (32) Lovell, C. S.; Walker, B. L.; Damion, R. A.; Radhi, A.; Tanner, S. F.; Budtova, T.; Ries, M. E. Biomacromolecules 2010, 11, 2927–2935. (33) Mason, P. E.; Neilson, G. W.; Enderby, J. E.; Saboungi, M. L.; Brady, J. W. J. Am. Chem. Soc. 2005, 127, 10991–10998. (34) Mason, P. E.; Neilson, G. W.; Enderby, J. E.; Saboungi, M. L.; Brady, J. W. J. Phys. Chem. B 2005, 109, 13104–13111. (35) Remsing, R. C.; Swatloski, R. P.; Rogers, R. D.; Moyna, G. Chem. Commun. 2006, 1271–1273. (36) Youngs, T. G. A.; Hardacre, C.; Holbrey, J. D. J. Phys. Chem. B 2007, 111, 13765–13774. (37) Remsing, R. C.; Hernandez, G.; Swatloski, R. P.; Massefski, W. W.; Rogers, R. D.; Moyna, G. J. Phys. Chem. B 2008, 112, 11071– 11078. (38) da Costa Sousa, L.; Chundawat, S. P. S; Balan, V.; Dale, B. E. Curr. Opin. Biotechnol. 2009, 20, 339–347. (39) Peng, X.; Ren, J.; Sun, R. Biomacromolecules 2010, 11, 3519– 3524. (40) Bergenstable, M.; Wrhlert, J.; Himmet, M. E.; Brady, J. W. Carbohydr. Res. 2010, 345, 2060–2066. (41) Jinxin, G.; Zhang, D.; Liu, C. J. Theor. Comput. Chem. 2010, 9, 611–624. (42) Janesko, B. D. Phys. Chem. Chem. Phys. 2011, 13, 11393–11401. (43) Mohamed, M. N. A.; Watts, H. D.; Guo, J.; Catchmark, J. M. Carbohydr. Res. 2010, 345, 1741–1751. (44) Jinxin, G.; Zhang, D.; Duan, C.; Chengbu, L. Carbohydr. Res. 2010, 345, 2201–2205. (45) Derecskei, B.; Derecskei-Kovacs, A. Mol. Sim. 2006, 32, 109– 115. (46) Liu, H.; Sale, K. L.; Holmes, B. M.; Simmons, B. A.; Singh, S. J. Phys. Chem. B 2010, 114, 4293–4301. (47) Novoselov, N. P.; Sashina, E. S.; Petrenko, V. E.; Zaborsky, M. Fibre Chem. 2007, 39, 153–158. (48) Binder, J. B.; Raines, R. T. Proc. Natl. Acad. Sci. 2010, 107, 4516–4521. (49) Zhang, J.; Zhang, H.; Wu, J.; He, J.; Xiang, J. Phys. Chem. Chem. Phys. 2010, 12, 1941–1947. (50) Hesse-Ertelt, S.; Heinze, T.; Kosan, B.; Schwikal, K.; Meister, F. Macromol. Symp. 2010, 294, 75–89. (51) Appell, M.; Strati, G.; Willett, J. L.; Momany, F. A. Carbohydr. Res. 2004, 339, 537–551. (52) Csonka, G. I.; French, A. D.; Johnson, G. P.; Stortz, C. A. J. Chem. Theor. Comput. 2009, 5, 679–692. (53) Jockusch, R. A.; Talbot, F. O.; Rogers, P. S.; Simone, M. I.; Fleet, G. W. J.; Simons, J. P. J. Am. Chem. Soc. 2006, 128, 16771–16777. (54) Kozmon, S.; Tvaroska, I. Collect. Czech. Chem. C 2006, 71, 1453–1469. (55) Momany, F. A.; Appell, M.; Willett, J. L.; Bosma, W. B. Carbohydr. Res. 2005, 340, 1638–1655. (56) Hutter, J.; Ballone, J. P.; Bernasconi, M.; Focher, P.; Fois, E.; Goedecker, S.; Marx, D.; Parrinello, M.; Tuckerman, M. E. CPMD; Max Planck Institut fuer Festkoerperforschung and IBM Zurich Research Laboratory: Stuttgart, 1990. (57) Frisch, M. J.; et al. Gaussian 03, revision C.02; Gaussian Inc.: Wallingford, CT, 2004. (58) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865–3868. (59) Troullier, N.; Martins, J. L. Phys. Rev. B 1991, 43, 1993. (60) Barnett, R. N.; Landman, U. Phys. Rev. B 1993, 48, 2081–2097. (61) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (62) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789. (63) Parr, R. G.; Wang, W. Density-Functional Theory of Atoms and Molecules; Oxford University Press: New York, 1989. (64) Zahn, S.; Bruns, G.; Thar, J.; Kirchner, B. Phys. Chem. Chem. Phys. 2008, 10, 6909–7048. (65) Tsuzuki, S.; Uchimaru, T.; Mikami, M. J. Phys. Chem. B 2009, 113, 5617–5621. 840

dx.doi.org/10.1021/jp207989w |J. Phys. Chem. B 2012, 116, 833–840